The above link states that “Basic Compression (LZ4) - Inline and post process” are included starter. However as soon as I enable post compression I get a license violation notice. In prism element it also shows the available features for starter and post compressions shows as yes:
Is post compression available in Starter?
Also the link states that lz4hc compression is not available for Starter but running the following command on a starter clusters show it is available and compressing data:
curator_cli display_data_reduction_report
Has something changed to make lz4hc available to starter?
Is there another document or link with a comprehensive list of features available for each edition?
Thanks,
--Zip
Page 1 / 1
hey @zip , that is correct Basic Compression (LZ4) - Inline and post-process are supported in Starter license.
Whereas I am pretty sure that Deep Compression (LZ4HC) - Greater efficiency for cold data is not supported for a Starter license.
I think the license violation issue must be a GUI issue or a bug will have to check with the Licensing team.
Thanks for the response. I have two follow up questions:
Is it correct to say that you can only have inline compression or post process compression but you cannot have both enabled a the same time.
Since Deep Compression(LZ4HC) works on cold data I assume it is done using post-process compression. Is it safe to assume that inline-compression does not use “Deep Compression(LZ4HC)?
Hey @zip , Both of your points are correct.
Inline compression - Data is compressed before it is written.
Post-process Compression- Once the data cools down it is then eligible to become compressed i.e after it passes the compression delay. All new write I/O are written in an uncompressed state. After the compression delay (configurable) is met, the data is eligible to become compressed in this scenario.
This is a great video which highlights the difference between both:
Normal data will be compressed using LZ4 which provides a very good blend between compression and performance. For cold data, LZ4HC will be leveraged to provide an improved compression ratio.
Hope it helps mate. Let me know if I can help in any other way.
Hey Anish,
Prior to posting here I watched that video and also read that link but was still uncertain.
Thanks to Nutanix’s excellent documentation I understand how inline compression and post compression work. I also understand that LZ4HC works on cold data.
If I have Compression delay set to zero as seen in this picture:
My question is: Is it correct to assume if delay is set to zero there will be NO post-process compression. In addition since LZ4HC is a post-compression process I assume there will be no LZ4HC compression either.
Thanks,
--Zip
Hey @zip yes, that is absolutely correct, i.e if the delay is set to zero there will be NO post-process compression and it will be INLINE compression.
Exactly, with INLINE compression, there will be no LZ4HC compression too.
Both your points are correct.
I hope I cleared your doubts. Let me know if I can help in any other way.